Skip to main content
ABCT 2023 Annual Convention Main banner
Final Program


Program Flipbook
Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.
Presentation Icons
Ticketed Session
Ticketed Session
Recorded Session
Recorded Session
Back
ViewAttendees 4

Symposia

Schizophrenia / Psychotic Disorders

Session: Digital health for schizophrenia-spectrum disorders: From local innovation to global impact

2 - (SYM 61) Bringing the Promise of Artificial Intelligence to a Community Behavioral Health Clinic near You: Introducing Cbtpro

Saturday, November 18, 2023
10:30 AM – 12:00 PM PST
Location: Chiwawa (501), Level 5

Cognitive Behavioral Therapy for psychosis (CBTp) is the most well-researched psychotherapy for psychotic disorders, yet fewer than 1% of American behavioral health practitioners are trained in the intervention (Kopelovich et al., 2022). Technology-enhanced learning holds promise for reshaping our approach to pre-service training, professional development, and clinical quality improvement.  We are engaged in a 2-phase fast-track NIMH-funded STTR (R42MH123215) to develop and evaluate CBTpro—a digital training tool that blends asynchronous multimodal training with Artificial Intelligence (AI) to support high-quality CBTp skills training. In Phase 1, we adhered to a Scrum Agile software development process, generated digital content, and developed a fidelity coding scheme based on discrete CBT skills. Tool validation was achieved through UCD interviews (N=21) and a 2-week field trial (N=20) among frontline practitioners in a community behavioral health clinic. Our iterative design-build process resulted in a platform with integrated digital content and deliberate practice opportunities with highly authentic patients (Maastricht assessment of Simulated Patients=8.4/10). Seven practice modules focus on building foundational CBT competencies (e.g., collaborative empiricism, Socratic dialogue) and seven modules target functional CBTp competencies (e.g., psychosis psychoeducation, reality testing). Learners engage in repeated practice of discrete CBTp skills with simulated patients and receive real-time AI-generated feedback. Machine learning model performance met or exceeded 80% of human reliability on a test set of responses across all modules. CBTpro was rated as a highly useful and navigable system (System Usability Scale mean score=86.5, range=70-100) with a high likelihood of being recommended to a colleague (X(SD)=9.4(0.8), range=8-10). In Phase 2, a randomized clinical trial examines both learner and clinical outcomes among behavioral health practitioners (N=100) and a sample of their outpatients with psychosis (N=300). We will demonstrate this first-of-its-kind functional AI training tool and present preliminary outcomes from the multi-site clustered RCT.